The effects of nitrous oxide (N2O) on cerebral blood flow and metabolism, intracranial pressure (ICP), the electroencephalogram etc. has been well described, at least when N2O is used alone. However, during neurosurgical procedures, N2O is almost always given in combination with either volatile or intravenous agents, and generally after the institution of some degree of hypocarbia. Unfortunately, the modifying influence of such interventions are not well known, and, therefore, the cerebral effects of 70% N2O were studied in rabbits anesthetized with either 1 MAC halothane or isoflurane, or with a fentanyl/pentobarbital combination, during both normocarbia (PaCO2 ~ 40 mm Hg) and hypocarbia (PaCO2 ~ 20 mm Hg). Cortical cerebral blood flow (CBFc) and sagittal sinus blood flow (CBFss - as an index of 'global' forebrain flow) were measured using the hydrogen clearance method. Cerebral oxygen consumption (CMRO2) was calculated, and intracranial pressure (ICP), central venous pressure, heart rate, mean arterial pressure, and the EEG were also recorded. CBFc during normocarbic halothane, isoflurane, and fentanyl-pentobarbital anesthesia was 69 ± 23, 41 ± 16, 53 ± 26 ml·100g-1·min-1 (mean ± SD) respectively, with a significant difference between halothane and isoflurane. The addition of 70% N2O to all three anesthetics significantly increased CBFc by 32%, 34% and 36% respectively during normocarbia and by 47%, 65% and 27% during hypocarbia. A similar pattern was seen for CBFss. There were no significant differences in the response to N2O based either on anesthetic or PaCO2, except that the increase in CBFss produced by N2O during normocarbic fentanyl-pentobarbital anesthesia (10%) was less than that noted during normocarbic halothane anesthesia (39%). CMRO2 was not changed by addition of N2O regardless of anesthetic or PaCO2. N2O increased ICP under all conditions; the increments were small (about 1 mm Hg). The greatest EEG changes in responses to N2O were seen in halothane-anesthetized animals (a decrease in amplitude and increase in frequency). In contrast, no consistent EEG changes were seen with N2O in the fentanyl/pentobarbital group, while the only effect observed in the presence of isoflurane was the disappearance of the typical burst-suppression pattern. We conclude that N2O increases CBF and ICP with each of the three background anesthetics during both normocarbia and hypocarbia. The magnitude of this response to N2O may be less during a fentanyl/barbiturate based anesthetic. These changes were not associated with alterations in CMRO2, suggesting that N2O may be a direct cerebral vasodilator in this species.